Project Summary Repetitive transcranial magnetic stimulation (rTMS) is an effective treatment for major depressive disorder, but remission rates are suboptimal and ideal stimulation parameters are unknown. rTMS is thought to work by changing the synaptic strength of neurons, and the ability of our brain to make these changes is referred to as plasticity. rTMS-induced changes become more pronounced with successive treatment sessions, a buildup referred to as metaplasticity. While both plasticity and metaplasticity are well-established in single cell physiology, relevance to rTMS in humans remains unknown. To improve clinical efficacy, we need to understand 1) the neural response to a single rTMS session (plasticity), 2) the neural response to repeated daily rTMS sessions (metaplasticity), and 3) whether computational models of plasticity based on single-cell physiology apply to human patients receiving rTMS for depression. This project tests the hypothesis that using TMS paired with EEG, neural changes that accumulate during treatment can predict clinical outcome using established computational models. I plan to test this hypothesis by randomizing depressed patients to four weeks of daily left prefrontal active or sham 10Hz rTMS. I will measure the single pulse TMS-evoked potential (TEP), a well-studied causal EEG measure of brain excitability, before, during, and after every rTMS session. TEPs consist of an early (<50ms) and late (50-250ms) response that likely reflect local excitation and inhibition, respectively. TEPs will be measured locally in the left lateral prefrontal cortex and compared to downstream sites in parietal and medial prefrontal cortex. I will then evaluate the relationship between brain state, plasticity, metaplasticity, and clinical improvement. Finally, I will test if these changes fit an established model of metaplasticity derived from single cell studies in animals. The current study will 1) establish a detailed mechanistic understanding of the electrophysiological effects of rTMS treatment; 2) identify clinically meaningful electrophysiological biomarkers for rTMS treatment; and 3) establish a computational model to help predict both brain and clinical changes. If successful, this project will provide the basis for novel stimulation protocols designed to maximize brain changes and clinical response. !